Later this decade, Volvo will introduce cars built on the Nvidia Drive Thor, which is capable of up to 1,000 trillion operations per second. That’s four times as many operations per second as the current Drive Orin SoC, while offering seven times greater energy efficiency.
06 Sep 2024 | 2646 Views | By Autocar Professional Bureau
Software-defined vehicles (SDVs) are a global automotive megatrend and for Volvo Cars, the new Volvo EX90 is the first Volvo car to be truly software-defined. The EX90 is built on a centralised core compute architecture, made possible through its long-standing collaboration with software major Nvidia.
The EX90’s industry-leading core ...
This is an Autocar Pro Plus article. Subscribe to continue reading.
Software-defined vehicles (SDVs) are a global automotive megatrend and for Volvo Cars, the new Volvo EX90 is the first Volvo car to be truly software-defined. The EX90 is built on a centralised core compute architecture, made possible through its long-standing collaboration with software major Nvidia.
The EX90’s industry-leading core computing system is powered by an Nvidia Drive Orin system-on-a-chip (SoC), which is capable of more than 250 trillion operations per second (TOPS). This core compute system orchestrates everything in the car: from powering the deep learning capabilities underpinning our AI-based active safety and driving-assistance systems to helping introduce safe autonomous driving in the future and delivering best-in-class customer experience.
Now, in an effort to achieve the full potential of its software-defined cars with a centralised core compute architecture, Volvo is taking its collaboration with Nvidia to the next level. Later this decade, we will introduce cars built on the Nvidia Drive Thor, which is capable of up to 1,000 TOPS. That’s four times as many operations per second as one Drive Orin SoC, while offering seven times greater energy efficiency.
Integrating Drive Thor will help further future-proof Volvo’s next generation of cars, states the Swedish carmaker. Drive Thor, which integrates the Nvidia Blackwell GPU architecture, will enable Volvo to deploy even more advanced driving-assistance and safety features, develop autonomous driving, and introduce generative AI-based capabilities and in-car experiences.
“With Nvidia Drive Thor in our future cars, our in-house-developed software becomes more scalable across our product line-up, which will help us to continue to improve the safety in our cars, deliver best-in-class customer experiences, reduce our costs and increase our margins,” says Jim Rowan, CEO of Volvo Cars.
To further explore the potential of AI, Volvo Cars, through its fully owned software company, Zenseact, is also using the Nvidia DGX systems – an AI supercomputing platform optimised for large workloads – to help develop safe autonomous driving.
The DGX systems for AI model training will be used before deploying to our future fleets on the road. With the power of the Nvidia DGX platform – utilising its purpose-built AI infrastructure and optimised software stack – we can enhance the efficiency of training both current and future AI models.
“The Nvidia DGX AI supercomputer will supercharge our AI training capabilities, making this in-house AI training data centre one of the largest in the Nordics,” says Anders Bell, Chief Engineering & Technology Officer at Volvo Cars. “By leveraging Nvidia technology and setting up the data centre, we can pave a quick path to high-performing AI, ultimately helping make our products safer and better.”